Hong Kong Physiotherapy Journal (2017) 37, 46e49

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.hkpj-online.com

RESEARCH PAPER

Chinese translation and validation of the

Oxford Knee Scale for patients with knee
osteoarthritis
Roy T.H. Cheung, PT, PhD a,*, Shirley P.C. Ngai, PT, PhD a,
Kevin K.W. Ho, MD b

a
Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hong Kong, China
b
Department of Orthopaedics and Traumatology, The Chinese University of Hong Kong, Hong Kong,
China

KEYWORDS Abstract Background: Oxford Knee Scale (OKS) is a commonly used instrument to assess the
elderly; symptoms and functional status in people with knee osteoarthritis. However, a Chinese version
outcome of this scale is not yet available.
measurement; Objective: The objective of this study was to translate the OKS into Chinese and validate the
questionnaire Chinese version of OKS.
Methods: The Chinese OKS was translated from the original English version following the recom-
mendations of the International Society for Pharmacoeconomics and Outcomes Research. One
hundred Chinese reading patients with knee osteoarthritis were recruited from local hospitals
and physiotherapy clinics. Psychometric properties were evaluated in terms of testeretest reli-
ability and internal consistency. Convergent validity was examined by Spearman rank correlation
coefficient tests by comparing its score with the validated Chinese version of the Western Ontar-
io and McMaster Universities Osteoarthritis Index and Health Outcome Survey Short Form-36.
Results: Chinese OKS demonstrated excellent reliability (intraclass correlation coeffi-
cient Z 0.88). Cronbach a of individual questions was > 0.7. Strong correlation was found be-
tween the Chinese OKS and the Western Ontario and McMaster Universities Osteoarthritis
Index (r > 0.553, p < 0.001). Fairly strong negative correlation was also found between Chinese
OKS and Health Outcome Survey Short Form-36 (r Z 0.273 to 0.666, p < 0.05).
Conclusion: The Chinese translated version of OKS is a reliable and valid instrument for clinical
evaluation in Chinese reading patients with knee osteoarthritis.
Copyright ª 2017, Hong Kong Physiotherapy Association. Published by Elsevier (Singapore) Pte
Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.
org/licenses/by-nc-nd/4.0/).

* Corresponding author. Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Gait & Motion Analysis Lab, ST004,
G/F, Block S, Hung Hom, Hong Kong, China.
E-mail address: Roy.Cheung@polyu.edu.hk (R.T.H. Cheung).

http://dx.doi.org/10.1016/j.hkpj.2017.03.002
1013-7025/Copyright ª 2017, Hong Kong Physiotherapy Association. Published by Elsevier (Singapore) Pte Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction
Knee osteoarthritis is a major healthcare concern. The
overall lifetime risk of symptomatic knee osteoarthritis is
an astounding 50% [1]. According to the statistics from the
World Health Organization [2], the current Chinese popu-
lation is > 1.4 billion. It has been suggested that up to 10.3%
of people in China suffer from symptomatic knee osteoar-
thritis [3]. Therefore, it is estimated that there are
currently > 100 million patients with knee osteoarthritis in
China. It is foreseeable that the number of Chinese reading
patients with knee osteoarthritis will increase with the
aging population, thereby increasing the already significant
economic burden.
For the purpose of evidence-based practice, reliable and
accurate clinical tools are required to document the prog-
ress and evaluate treatment response for clinical decision-
making. The Oxford Knee Scale (OKS) was introduced to
assess the symptoms and functional status in patients with
knee osteoarthritis [4]. OKS has 12 items addressing the
extent of knee pain and the level of functional impairments
related to daily activities over the previous 4 weeks. This
scale has been used worldwide for evaluating patients
before and after total knee arthroplasty and has been
translated in other languages [5e10].
To date, there is no validated Chinese version of this
outcome measurement. Hence, the purpose of this study
was to test the reliability, internal consistency, and
convergent validity of the Chinese translation of OKS.
Methods
Participants
A total of 100 patients (35 men and 65 women) with a
diagnosis of knee osteoarthritis were recruited from the
orthopedic outpatient department and physiotherapy
clinics of local hospitals in Hong Kong, China. The diagnosis
was confirmed by radiographic findings by an orthopedic
surgeon and they were able to read and comprehend Chi-
nese. We excluded patients if they had corticosteroid in-
jection or serotonin treatment within the previous 8
weeks, or another site of osteoarthritis other than the
tibiofemoral joint. We also excluded patients who were
illiterate. The demographical data (age, height, weight,
and body mass index) and information about knee osteo-
arthritis (distribution of affected leg and duration of knee
osteoarthritis) of the patients are presented in Table 1.
The study protocol was reviewed and approved by the
ethical committees of the involved university and hospi-
tals. All the patients provided their written informed con-
sent before being tested.
Development of the Chinese questionnaire
The translation process followed the procedures recom-
mended by the International Society for Pharmacoeco-
nomics and Outcomes Research [11] and other crosscultural
adaptation and translation studies [12]. In brief, two Chi-
nese translations of the original English version of OKS
were performed by two independent EnglisheChinese
47
Table 1 Demographics of participants
Demographics Values (SD)
Age (y) 65.6 (8.9)
Height (m) 1.59 (0.09)
Weight (kg) 69.1 (16.6)
Body mass index (kg/m2) 27.2 (4.2)
Involved knee Left 21
Right 10
Both 69
Duration of knee osteoarthritis (y) 7.9 (1.0)
SD Z standard deviation.
translators. A panel comprising a translator and two bilin-
gual authors compared the two translations and formulated
a consensus version. Backward translation of the consensus
version questionnaire to English version was performed by
two independent translators who were blinded to the
original questionnaire and not involved in the previous
translation work. The backward translated questionnaire
was reviewed for equivalence to the original questionnaire
by an expert panel, which consisted of two experienced
physiotherapists and two research fellows in the area of
musculoskeletal physical therapy and orthopedics. The
questionnaire was finalized after a pilot test for cross-
cultural adaptation, which involved 20 patients with knee
osteoarthritis.
Statistical analysis
Testeretest reliability of the Chinese OKS was determined
by comparing the scores obtained from two subsequent
treatment sessions (within 7 days apart) by the intraclass
correlation coefficient. Internal consistency of the ques-
tionnaire was examined by the Cronbach a. Based on pre-
vious validation studies [12,13], convergent validity was
assessed by comparing the scores with the validated Chi-
nese versions of the Western Ontario and McMaster Uni-
versities (WOMAC) Osteoarthritis Index [14] and Health
Outcome Survey Short Form-36 (SF-36) [15] by the Spear-
man’s rank correlation coefficients (r). The correlation
value was considered to be very strong if it was between 0.9
and 1.0, strong if it was between 0.7 and 0.9, moderate if it
was within 0.5e0.7, and weak if it was < 0.5 [16].
Results
Crosscultural adaptation and psychometric
properties
We did not receive any critique about the Chinese OKS in
the pilot test involving 20 patients with knee osteoarthritis.
All the participants expressed that they understood the
wordings used in the instrument. Therefore, the question-
naire was used in the subsequent validation study without
any further adaptation. Thirty individuals were asked to fill
in the Chinese OKS twice on subsequent two clinical
visits (within 7 days apart). The testeretest reliability
was excellent (intraclass correlation coefficient Z 0.88,
48 R.T.H. Cheung et al.

p < 0.001). The Cronbach a of the total score in the Chinese

Table 3 Spearman rank correlation coefficients between
OKS was 0.802. Cronbach a when each question was
OKS and WOMAC Osteoarthritis Index
deleted ranged from 0.712 to 0.855 (Table 2).
WOMAC Osteoarthritis Index OKS

Convergent validity Total score 0.832*

Subcategory e pain 0.846*
Strong correlation was found between the Chinese OKS and Subcategory e stiffness 0.553*
the total score of WOMAC Osteoarthritis Index (r Z 0.832, Subcategory e physical function 0.793*
p < 0.001). The score in Chinese OKS was also associated OKS Z Oxford Knee Scale; WOMAC Z Western Ontario and
with the subcategories in the WOMAC Osteoarthritis Index McMaster Universities.
(Table 3). The strength of correlation was highest in the * Significant correlation with p < 0.001.
pain domain (r Z 0.846, p < 0.001), followed by physical
function (r Z 0.793, p < 0.001). The weakest association
The WOMAC Osteoarthritis Index and SF-36 were chosen
was found between the joint stiffness domain (r Z 0.553,
to test the convergent validity as these two instruments
p < 0.001) but the strength of the correlation was not weak.
were shown to be valid and highly reliable [14,15]. The
The correlations between Chinese OKS and SF-36 were
strong correlations of between the WOMAC Osteoarthritis
diverse (Table 4). The strongest correlations were between
Index suggested strong agreement between the two
bodily pain (r Z 0.666, p < 0.001), physical functioning (r
different scales and indicated convergent validity of the
Z 0.530, p < 0.001), and role-physical (r Z 0.483,
Chinese OKS. Since SF-36 assesses generic health instead
p < 0.001). A weaker correlation was also noticed between
of disease/pathology-specific parameters, a weaker cor-
the Chinese OKS and nonphysical domains, such as social
relation (r range from 0.273 to 666) between the
functioning (r Z 0.456, p < 0.001) and mental health (r
Chinese OKS and SF-36 was expected. When comparing the
Z 0.412, p < 0.001). The relationship between the Chi-
OKS (with only 12 questions,) patients are required to
nese OKS and the general health was the weakest (r Z
spend more time to administer WOMAC Osteoarthritis
0.273, p < 0.05).
Index (with 24 questions) or SF-36 (with 36 questions).
Therefore, the application of OKS should be more clini-
cally friendly.
Discussion
Two major limitations should be considered in the pre-
sent study. First, since OKS was not only used to evaluate
The Chinese OKS demonstrated acceptable psychometric
patients with knee osteoarthritis but also patients who
properties in the patients with knee osteoarthritis. The
received total knee arthroplasty. Testing of the Chinese
findings of the current study suggested that this instru-
OKS in other patient cohorts in the future is warranted.
ment is a reliable and valid outcome measurement for
Second, previous validation studies usually had two to 20
the selected patient group in the Chinese reading
samples per item [17]; while we had 100 patients for the 12-
population.
item OKS, which may not be convincingly adequate.
Excellent testeretest reliability and internal consistency
of the Chinese OKS was demonstrated in the patients with
knee osteoarthritis. Such a finding was in accordance with Conclusion
other validation studies in similar patient cohorts [12,14],
which could be explained by the disease chronicity of our The Chinese translated version of OKS is a reliable and valid
sample. In addition, the OKS addresses some common sce- instrument for patients with knee osteoarthritis. Local
narios in everyday life and therefore no cultural modifica- research and multinational studies can be facilitated by the
tion of terms and wordings was required. current study.

Table 4 Spearman rank correlation coefficients of the

Table 2 Cronbach a if each question was deleted total score of OKS and SF-36
Question no. Cronbach a if item deleted SF-36 OKS
1 0.813 Physical functioning 0.530**
2 0.830 Role-physical 0.483**
3 0.816 Bodily pain 0.666**
4 0.855 General health 0.273*
5 0.818 Energy vitality 0.428**
6 0.812 Social functioning 0.456**
7 0.826 Role-emotional 0.358**
8 0.824 Mental health 0.412**
9 0.712
10 0.833 OKS Z Oxford Knee Scale; SF-36 Z Health Outcome Survey
Short Form-36.
11 0.818
* Significant correlation with p < 0.05.
12 0.809
** Significant correlation with p < 0.001.

Conflicts of interest
None.
Funding/support
This study received no funding support.
Authors’ contributions
Dr. Roy Cheung: Study design; Subject recruitment; Data
collection; Data analyses; Data interpretation; Project
management; Writing/revising the manuscript.
Dr. Shirley Ngai: Data analyses; Data interpretation;
Writing/revising the manuscript.
Dr. Kevin Ho: Subject recruitment; Data collection; Data
analyses; Data interpretation; Writing/revising the
manuscript.
Acknowledgments
The authors thank the contribution from Mr Chris Kwok for
his contribution for data collection in this project.
References
[1] Murphy L, Schwartz TA, Helmick CG, Renner JB, Tudor G,
Koch G, et al. Lifetime risk of symptomatic knee osteoar-
thritis. Arthritis Rheum 2008;59:1207e13.
[2] World Health Organization. China. Available from: http://
www.who.int/countries/chn/en/. [accessed August 29, 2016].
[3] Tang X, Wang S, Zhan S, Niu J, Tao K, Zhang Y, et al. The
prevalence of symptomatic knee osteoarthritis in China: re-
sults from the China Health and Retirement Longitudinal
Study. Arthritis Rheumatol Hoboken NJ 2016;68:648e53.
[4] Dawson J, Fitzpatrick R, Murray D, Carr A. Questionnaire on
the perceptions of patients about total knee replacement.
J Bone Joint Surg Br 1998;80:63e9.
[5] Dunbar MJ, Robertsson O, Ryd L, Lidgren L. Translation and
validation of the Oxford-12 item knee score for use in Sweden.
Acta Orthop Scand 2000;71:268e74.
[6] Charoencholvanich K, Pongcharoen B. Oxford knee score and
SF-36: translation & reliability for use with total knee
49
arthroscopy patients in Thailand. J Med Assoc Thail Chotmai-
het Thangphaet 2005;88:1194e202.
[7] Takeuchi R, Sawaguchi T, Nakamura N, Ishikawa H, Saito T,
Goldhahn S. Cross-cultural adaptation and validation of the
Oxford 12-item knee score in Japanese. Arch Orthop Trauma
Surg 2011;131:247e54.
[8] Xie F, Li S-C, Roos EM, Fong K-Y, Lo N-N, Yeo S-J, et al. Cross-
cultural adaptation and validation of Singapore English and
Chinese versions of the Knee injury and Osteoarthritis
Outcome Score (KOOS) in Asians with knee osteoarthritis in
Singapore. Osteoarthr Cartil OARS Osteoarthr Res Soc 2006;14:
1098e103.
[9] Haverkamp D, Breugem SJM, Sierevelt IN, Blankevoort L, van
Dijk CN. Translation and validation of the Dutch version of the
Oxford 12-item knee questionnaire for knee arthroplasty. Acta
Orthop 2005;76:347e52.
[10] Gonçalves RS, Tomás AM, Martins DI. Cross-cultural adaptation
and validation of the Portuguese version of the Oxford Knee
Score (OKS). The Knee 2012;19:344e7.
[11] Wild D, Grove A, Martin M, Eremenco S, McElroy S, Verjee-
Lorenz A, et al. Principles of Good Practice for the Translation
and Cultural Adaptation Process for Patient-Reported Out-
comes (PRO) Measures: report of the ISPOR Task Force for
Translation and Cultural Adaptation. Value Health J Int Soc
Pharmacoeconomics Outcomes Res 2005;8:94e104.
[12] Cheung RTH, Ngai SPC, Ho KKW. Chinese adaptation and
validation of the Knee Injury and Osteoarthritis Outcome
Score (KOOS) in patients with knee osteoarthritis. Rheumatol
Int 2016;36:1449e54.
[13] Cheung RTH, Ngai SPC, Lam PL, Chiu JKW, Fung EYH. Chinese
translation and validation of the Kujala scale for patients with
patellofemoral pain. Disabil Rehabil 2012;34:510e3.
[14] Xie F, Li S-C, Goeree R, Tarride J-E, O’Reilly D, Lo N-N, et al.
Validation of Chinese Western Ontario and McMaster Univer-
sities Osteoarthritis Index (WOMAC) in patients scheduled for
total knee replacement. Qual Life Res Int J Qual Life Asp Treat
Care Rehabil 2008;17:595e601.
[15] Lam CL, Gandek B, Ren XS, Chan MS. Tests of scaling as-
sumptions and construct validity of the Chinese (HK) version
of the SF-36 Health Survey. J Clin Epidemiol 1998;51:1139e47.
[16] Hinkle DE, Wiersma W, Jurs SG. Applied statistics for the
behavioral sciences. 5th ed. Boston: Houghton-Mifflin; 2002.
[17] Arafat SMY, Chowdhury HR, Qusar MS, Hafez MA. Cross cultural
adaptation & psychometric validation of research in-
struments: a methodological review. J Behav Health 2016;5:
129e36.